mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: rainer.koenig.abg@t-online.de (Rainer Koenig)
To: greg@bakers.org
Cc: linux-kernel@vger.kernel.org
Subject: Re: 2.4.26 agpgart on Tyan K8W 32bit not working?
Date: Tue, 25 May 2004 21:38:45 +0200	[thread overview]
Message-ID: <87y8nguwje.fsf@Rainer.Koenig.Abg.dialin.t-online.de> (raw)
In-Reply-To: <1ZPD4-17S-7@gated-at.bofh.it> (bakerg3@yahoo.com's message of "Tue, 25 May 2004 21:20:10 +0200")

Hi,

gb <bakerg3@yahoo.com> writes:

> I'm using the generic 2.4.26 kernel on a Tyan K8W
> (dual opteron) with 4GB of memory and an NV18GL
> [Quadro4 NVS] graphics card.  The K8W BIOS version is
> 2.02 (1.02 also exhibits the same behavior).

Let me limit the output to the interesting part:

> On a 64bit kernel, the following is returned:
>
> [greg@testpig3 greg]$ lspci
> 04:01.0 PCI bridge: Advanced Micro Devices [AMD]
> AMD-8151 AGP Bridge (rev 13)
> 05:00.0 VGA compatible controller: nVidia Corporation
> NV18GL [Quadro4 NVS] (rev a2)
> 05:00.0 VGA compatible controller: nVidia Corporation
> NV18GL [Quadro4 NVS] (rev a2)
>
> On a 32-bit kernel, the following is returned:
>
> [greg@profit linux]$ lspci
> 03:0c.0 FireWire (IEEE 1394): Texas Instruments
> TSB43AB22/A IEEE-1394a-2000 Controller (PHY/Link)
> 05:00.0 VGA compatible controller: nVidia Corporation
> NV18GL [Quadro4 NVS] (rev a2)

The AMD-8151 AGP Bridge on bus #4 is missing on the 32-bit kernel.

> [greg@profit linux]$ dmesg | grep -i agp
> Linux agpgart interface v0.99 (c) Jeff Hartmann
> agpgart: Maximum main memory to use for agp memory:
> 3932M
> agpgart: no supported devices found.
> [drm:drm_init] *ERROR* Cannot initialize the agpgart
> module.
> 0: NVRM: AGPGART: unable to retrieve symbol table

Well, I have experienced problems with Red Hat EL 3.0 32 bit where the
old kernel doesn't support the AMD bridge. But on my tests a 2.4.25
kernel did well. So the problem is located elsewhere.

> I have tried booting with:
>
> kernel /vmlinuz-2.4.26-bangalore4-01 ro root=/dev/hda2
> hdc=ide-scsi apm=power-off agp_try_unsupported=1

The question is how your kernel is configuring the PCI devices because
the problem is that the bridge is not listed in the output of lspci.
Try the boot parameter "pci=noacpi" so that PCI is configured the "old"
way without using probably buggy ACPI tables from the BIOS. Anyway, your
problem really looks very strange to me, because usally the kernel that
doesn't see the bridge on bus #4 won't even see the AGP card on bus #5.

Regards,
Rainer
-- 
Rainer König, Diplom-Informatiker (FH), Augsburg, Germany

       reply	other threads:[~2004-05-25 19:40 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <1ZPD4-17S-7@gated-at.bofh.it>
2004-05-25 19:38 ` Rainer Koenig [this message]
2004-05-25 19:21 gb
  -- strict thread matches above, loose matches on Subject: below --
2004-05-25 19:12 gb

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=87y8nguwje.fsf@Rainer.Koenig.Abg.dialin.t-online.de \
    --to=rainer.koenig.abg@t-online.de \
    --cc=greg@bakers.org \
    --cc=linux-kernel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®